Since time immemorial, physicians have gained insight into tissue biology through diagnostic palpation, the physical examination technique by which mechanical tissue property changes are detected. Changes in tissue mechanics typically accompany common disease processes, including fibrosis, inflammation, and neovascularization. These changes can be assessed with new advanced ultrasound techniques, termed ultrasound elastography.

Elastography Physics

Elastography is the set of techniques by which tissue stiffness is estimated as a physical property termed the Young’s modulus (E). The Young’s modulus is a proportionality constant that relates applied force per unit area or stress, and the resultant relative change in tissue dimension, or strain. Ultrasound elastography methods may be divided into two categories: quasi-static, or strain based, and dynamic, or shear wave based.

The nature of the external mechanical stimulus defines these methods. In strain-based elastography, force is applied by the application of probe pressure or through endogenous mechanical force (e.g., carotid pulsation). In shear wave-based elastography, a tissue shear wave is induced by the imaging system. In both approaches, the response of tissue to these mechanical stimuli is used to estimate tissue mechanical properties. Strain imaging uses the direct relationship E = \(\sigma /\varepsilon\) (Hooke’s Law) in which \(\sigma\) represents externally applied stress, and \(\varepsilon\) represents strain [1, 2]. Young’s modulus is usually not computed with clinical strain imaging systems, as the applied force on the tissue of interest is usually not known. Shear wave imaging systems compute Young’s modulus using the relationship E = 3ρc 2s in which \(\rho\) represents tissue density, and cs represents shear wave speed [1, 2]. Most vendors provide automatic calculation systems convert kPa to m/s and m/s to kPa and most ultrasound systems can display a table that indicates stiffness values both in kPa and m/s.

Strain elastography (SE)

SE measures tissue stiffness by applying external tissue pressure [3]. Tissue dimensions change due to the applied pressure; this deformation is termed strain. Stiffer lesions deform less, and have correspondingly lower strain and higher Young’s modulus. The strain ratio can be computed as the ratio between strain in a region of tissue and strain in a reference region of tissue. Computation of the strain ratio does not require knowledge of the applied force. For this reason, strain ratio is commonly used in clinical practice, and is mathematically equivalent to the Young’s modulus ratio between two tissue regions, assuming applied force is equivalent across these regions.

Strain elastography can be further divided into two groups by the method of tissue excitation (external manual excitation or excitation with internal physiological movement) [1]. Excitation with manual pressure measures elasticity in superficial tissues. A disadvantage of this excitation method is that manual stress is not efficiently transmitted to deeper tissues. Excitation from natural physiologic motion, such as cardiac pulsation and respiration, is another mechanism of generating tissue stress. Deep organs can be assessed with this method [1]. A variety of strain elastography implementations are available on clinical ultrasound systems, including ElaXtoTM, Real-time tissue elastography™, ElastoScan™, eSie Touch™, and Elasticity Imaging by the manufacturers Esaote, Hitachi, GE, Philips, Toshiba, Ultrasonix, Mindray, Samsung, and Siemens [1, 2]. A strain image example, in comparison with a conventional ultrasound image, is indicated in Fig. 1. In Virtual Touch™ Imaging (VTI), strain elastography is performed with the help of an acoustic push pulse, eliminating the need for an external/internal excitation method [2].

Fig. 1
figure 1

Palpable breast mass from a 24-year-old woman proven to be a benign fibroadenoma. A conventional B-mode image on the left, and a map of relative tissue stiffness in the same region of interest on the right. On the elastogram, bright areas depict tissue that is less stiff than tissue in the dark areas. Images were acquired using a L9 probe on a Siemens S2000 US system with manual strain (Courtesy of Dr. Richard Barr, MD, PhD)

Full size image

In strain imaging, tissue displacement is calculated by processing radiofrequency (RF) datasets obtained before and after compression [4]. Translucent colored elastograms (strain images) can be superimposed on B-mode images to provide complementary anatomic information. It is common to display the strain map as colored pixels on a red/blue scale or gray scale [5]. Unfortunately, intermanufacturer display scale variability is substantial, limiting inter-vendor comparability of strain elastography images.

Parameters commonly used in strain elastography include:

  • Strain ratio measures tissue deformation compared between two regions of interest (ROI). Strain ratio > 1, is an indicator of relatively low strain high stiffness [2].

  • Elasticity scores or grading systems are qualitative systems that have been used in a wide spectrum of disease processes, including breast imaging, to assess lesions [5]. These systems typically classify elastography patterns in a range between benign and malignant [1, 5].

  • Fat-to-lesion strain ratio is the strain ratio between fat and a lesion [6].

  • Elastography-to-B-mode size ratio is an index of the maximum size of a lesion on elastogram to that on a corresponding B-mode image [7].

Generally, in deep organs such as the liver and kidney, tissue stress is obtained with the help of cardiac/arterial or respiratory motion. In superficial organs like the thyroid, tissue stress is obtained with the help of manual compression. A strain image example of a lesion in comparison with a conventional ultrasound image is presented in Fig. 1.

Shear wave elastography (SWE)

The compressive acoustic waves used for conventional B-mode image generation travel at high speeds through soft tissue (1450–1550 m/s). By contrast, mechanical shear waves used for shear wave elastography travel relatively slowly (1–10 m/s). Shear wave propagation velocity depends on tissue stiffness [2, 4]. In commercially available shear wave elastography systems, compressive acoustic waves are used both to induce and track shear waves. Acoustically induced shear waves travel perpendicular to compressive waves; tissue motion induced by these shear waves is monitored at multiple locations along the ultrasound probe, permitting shear wave velocity estimation [4]. Young’s modulus can be algebraically derived from the shear wave speed (SWS).

SWE can be used in many different tissues for a variety of applications including hepatic lesion characterization [8], renal lesion characterization [9], diffuse liver and renal disease evaluation [10, 11], breast mass diagnosis [12, 13], prostate cancer detection [14], thyroid lesion characterization [15], and tendon imaging [16].

Transient elastography (TE)

A low-frequency (50 Hz) mechanical push is generated by a mechanical actuator and a resulting shear wave is generated and evaluated [17, 18]. With this technique, parameters like anisotropy, viscosity or elastic non-linearity can also be obtained [19]. Shear wave propagation velocity is proportional to tissue stiffness, which increases with fibrosis [20]. 1D transient elastography is marketed under the trade name FibroScan®. TE measures tissue stiffness over a 1 cm diameter–4 cm length region of tissue, which is 100 times larger than that evaluated with liver biopsy. If the pulse is not transmitted and recorded successfully, the software does not provide a reading [21]. Stiffness values are presented in kPa. Controlled Attenuation Parameter (CAP), is a technology that quantifies hepatic steatosis by measuring the energy loss as the sound wave passes through the medium. Total attenuation at 3.5 MHz is expressed in dB/m and steatosis is estimated using the same radiofrequency data as elastography, in the same location that stiffness is measured [22] (Fig. 2).

Fig. 2
figure 2

Transient elastography acquisition on a phantom. A Time Motion (TM) mode B Amplitude (A) mode. TM and A modes are used to locate ideal liver part. C Shear wave propagation image. y-axis is distance from skin, x-axis is time. Slope of the dashed line is shear wave speed (Vs) [23]. Tissue stiffness value is indicated in kPa. In the left panel, controlled attenuation parameter (CAP) value, which quantifies steatosis level, is indicated in dB/m

Full size image

Point shear wave elastography (pSWE)

Focused ultrasound results in focal tissue displacement, a process termed acoustic radiation force impulse (ARFI) imaging. The resultant shear waves are tracked, yielding a shear wave speed estimate that is an algebraic function of tissue stiffness. Point SWE is available on the Siemens Virtual Touch™ Quantification (VTQ/ARFI) system and on the Philips ElastPQ™ system [2]. An example of pSWE application in a phantom is presented in Fig. 3.

Fig. 3
figure 3

pSWE acquisition in a phantom. Green box is the focus of ARFI excitation. Shear wave speed value is indicated in left panel

Full size image

2D shear wave elastography

In this technique, acoustic radiation force is used to displace tissue at multiple points. The resultant shear wave front is readily detectable with high frame rate imaging, which is used to monitor propagation of the shear waves in real time at multiple points in the image [24]. A quantitative elasticity image (elastogram) is presented as a colorized display map, with quantitative results available as shear wave propagation speed in m/s or as the algebraically derived Young’s modulus in kPa [25]. Real-time tissue stiffness color maps added to the B-mode image allows the operator to avoid confounding anatomic structures such as blood vessels [26]. Maximum elastogram sizes are 2–3 cm in side length with a linear probe, and 9 × 4 cm with a convex probe [24]. This technique is available on multiple ultrasound systems including Virtual Touch™ Imaging Quantification (VTIQ/ARFI) by Siemens, Shear Wave Elastography by Philips, Shear Wave™ Elastography by SuperSonic Imagine, 2D-SWE by GE Healthcare, and Acoustic Structure Quantification™ (ASQ) by Toshiba [2]. An example of 2D-SWE  in a phantom is presented in Fig. 4.

Fig. 4
figure 4

2D-SWE acquisition in a phantom. Blue box denotes elastographic field of view (FOV) and circle decodes region of interest. Tissue stiffness in kPa is indicated at the bottom of the image. The color scale can be adjusted by the user. Blue areas are less stiff than red areas

Full size image

Clinical applications

Liver

Strain elastography

The liver fibrosis index (LFI) [27, 28] has been shown to be an accurate technique to distinguish fibrosis stages with AUROC (Area under receiver operating characteristic) values of 0.82 for fibrosis stage F0–1 vs. F2–4 and 0.87 for fibrosis stage F0–3 vs. F4 [27]. Koizumi et al. reported that a different strain parameter, termed ‘elastic ratio’ (the strain distribution value, intrahepatic venous small vessels/value in the hepatic parenchyma), was highly correlated with biopsy-proven fibrosis stage (Spearman correlation 0.82) with AUROC values to diagnose F ≥ 2 (0.89), F ≥ 3 (0.94), and F = 4 (0.95) [29]. SE can also be used to evaluate liver masses, with significant differences reported between benign and malignant lesions [30].

Transient elastography

Transient elastography can be used in the diagnosis of liver fibrosis due to multiple etiological factors including chronic viral infection and excessive alcohol intake [31,32,33]. In a meta-analysis including chronic liver disease due to multiple etiological factors, TE showed summary sensitivity and specificity values of 0.79 (95% CI 0.74–0.82) and 0.78 (95% CI 0.72–0.83) for F2 stage and 0.83 (95% CI 0.79–0.86) and 0.89 (95% CI 0.87–0.91) for cirrhosis [34]. TE does not use B-mode anatomic imaging to define the tissue from which stiffness information is obtained. The operator uses A-Mode US to define a measurement location away from vascular structures [2]. Therefore, an operator is not able to select the same liver region for serial TE measurements over time. Obtaining reliable acquisitions requires (1) at least 10 valid measurements, (2) valid measurements/total measurements ratio ≥ 60%, and (3) interquartile range (IQR) less than 30% of median value. A short training period is typically required for TE operators [2, 35]. TE has been shown to be reproducible with inter-operator intraclass correlation coefficient (ICC) of 0.98 and intra-operator ICC of 0.98 [36]. CAP is an integrated technology which can be used simultaneously with liver fibrosis quantification in Fibroscan system. An image example of transient elastography is depicted in Fig. 5.

Fig. 5
figure 5

Liver elastography image examples: (1) 2D-SWE acquisition of liver with Supersonic Aixplorer. Color-coded elastogram with color scale on right top. SWE values are indicated below the scale. (2) pSWE acquisition of liver with Siemens ACUSON S3000. ARFI-induced technique measures SWS in the center area. (3) Transient Elastography measurement example with Fibroscan

Full size image

pSWE

pSWE can be used in HBV, HCV, hepatic toxicity, alcoholic liver disease, and autoimmune hepatitis-related liver fibrosis [37,38,39]. It can also be used as a screening tool for early stage fibrosis detection [40], although evidence of benefit in this setting is limited. pSWE has been shown to be useful in detection of liver fibrosis. In a recent meta-analysis of 23 studies, Hu et al. reported AUROC values to distinguish liver fibrosis stages ranging from 0.649 to 0.934 for F ≥ 2, 0.848 to 0.97 for F ≥ 3, and 0.723 to 0.98 for F4 [41]. Using point SWE technique on both the liver and spleen has greater discriminative power than assessment of the liver alone [42]. Setting the ROI away from the liver capsule is recommended, as this choice results in more reliable shear wave speed values [43]. pSWE is a reproducible and reliable liver stiffness assessment technique, with ICC values of 0.89 (95% CI 0.85–0.92) for intra-observer and 0.85 (95% CI 0.76–0.90) for inter-observer agreement [44]. An image example of pSWE is demonstrated in Fig. 5.

2D-SWE

2D-SWE is a useful and feasible technique for fibrosis staging in both pediatric and adult patients [10, 45]. 2D-SWE has good performance for fibrosis staging. For example, for the diagnosis of fibrosis stage F ≥ 2, AUROC of 0.862 has been reported, and for early cirrhosis diagnosis, AUROC of 0.926 has been reported [46]. Using a cut-off value 7.29 kPa, this technique reaches a sensitivity of 95.4% for fibrosis stage ≥ 2 [47]. Although the diagnostic ability of the techniques is similar, shear wave speed values obtained from 2D-SWE may show higher stiffness values [48]. An image example of 2D-SWE is depicted in Fig. 5.

Kidney

Strain elastography

Chronic kidney disease (CKD) patients have been reported to show higher strain index (ratio) values when compared to healthy volunteers [49]. SE can also be used to detect renal graft interstitial fibrosis, a manifestation of organ rejection as a long-term complication of renal transplantation. Early diagnosis of graft fibrosis may play a useful role in treatment decisions concerning immunosuppressive agents [50].

pSWE

Renal fibrosis and diabetic renal disease can be evaluated with pSWE techniques. pSWE can detect renal fibrosis with a sensitivity of 86.3% and specificity of 83.3% [51]. Yu et al. reported a correlation of 0.773 between urinary albumin-to-creatinine ratio (diabetic kidney disease marker) and shear wave speed determined by VTQ, implying pSWE may serve as a marker for diabetic kidney disease [52]. Renal pSWE has been shown to be moderately reliable with intraclass correlation coefficient (ICC) values of 0.71 in the right kidney and 0.69 in the left kidney [53]. Age and gender can influence the SWS measurements acquired from the kidney [54], 55].

2D-SWE

2D-SWE can be used to diagnose chronic kidney disease. CKD patients show higher stiffness values [9.4 kPa] when compared with healthy volunteers [4.4 kPa] [11]. 2D-SWE can also be used in the diagnosis of diabetic kidney disease (DKD). Hassan et al. reported a significant difference in cortical stiffness values of DKD patients and healthy subjects (23.7 kPa vs. 9.02). Furthermore, significant differences between CKD grades have been reported [56]. An image example of 2D-SWE in renal tissue is demonstrated in Fig. 6.

Fig. 6
figure 6

2D-SWE image of kidney. Mean stiffness is 10.4 kPa for this patient, likely reflecting elevated renal stiffness due to CKD-related fibrosis [11]

Full size image

Breast

Strain elastography

Several strain elastography features have been proposed, including strain ratio, elasticity score (Tsukuba score), and elastography-to-B-mode size ratio [57]. In a meta-analysis with 25 studies focusing on elasticity score and strain ratio, overall mean sensitivity and specificity values to  diagnose malignant breast lesions were reported as 0.834 (95% CI 0.814–0.853) and 0.842 (95% CI 0.829–0.854), respectively, for elasticity score and 0.883 (95% CI 0.844–0.916) and 0.814 (95% CI 0.786–0.839) respectively, for strain ratio [58]. Furthermore, tumor grade may also be distinguished using an elasticity imaging/B-mode ratio [59]. It has been shown that the addition of strain elastography to a conventional gray scale ultrasound-based classification system—Breast Imaging Reporting and Data System (BI-RADS)—yields an AUROC of 0.875 for cancer detection with the ability to characterize lesions < 2 cm [60]. Breast lesion size prediction has been reported to be more accurate on elastographic images than conventional gray scale images when compared with the reference standard of the surgical excision specimen [61].

pSWE

Li et al. reported the diagnostic performance of pSWE to differentiate malignant and benign lesions in a meta-analysis of 11 studies, finding an overall sensitivity of 0.84 (95% CI 0.81–0.87) and specificity of 0.94 (95% CI 0.91–0.94) [62]. 2D-SWE and pSWE had similar performance to detect malignancy in breast tissue. In a different meta-analysis of 9 studies comparing 2D-SWE and pSWE, overall sensitivity and specificity values were reported as 0.91 (95% CI 0.88–0.94) and 0.82 (95% CI 0.75–0.87) for 2D-SWE and 0.89 (95% CI 0.81–0.94) and 0.91 (95% CI 0.84–0.95) for pSWE, respectively [63].

2D-SWE

2D-SWE is useful for differentiating benign and malignant breast lesions with reported AUROC values ranging from 0.74 to 0.98 [64]. The addition of 2D-SWE to conventional B-mode ultrasound can improve diagnostic performance by reducing the need for follow-up exams of patients with BI-RADS 3 breast lesions [61].

Prostate

Conventional screening and diagnostic methods for prostate evaluation include Prostate-Specific antigen (PSA) assessment, digital rectal exam (DRE), and transrectal ultrasound (TRUS)-guided biopsy.  However, the limited capacity of these techniques to accurately localize prostate cancer has led to the emergence of elastography as a tumor localization technique [65]. Elastographic techniques can be used to assess both benign prostatic hyperplasia (BPH) and prostate cancer (PC).

Strain elastography

In strain elastography, images are obtained with slight transrectal manual compression. An inflatable endorectal balloon may be used to generate endorectal prostate elastography images [61]. Although SE-guided prostate biopsy shows higher sensitivity when compared to conventional gray scale US-guided biopsy to detect prostate cancer (60.8% vs. 15%, respectively), only relying on SE results is not recommended [66]. Strain elastography has been shown to have a sensitivity of 58.8% and specificity of 43.3% to identify the prostate cancer index lesion (the main lesion responsible for possible metastasis) [61, 67]. An image example of strain elastography for prostate cancer diagnosis is shown in Fig. 7.

Fig. 7
figure 7

Transrectal strain image of prostate with distribution of tissue deformation as a colorized map. Blue represents relatively hard tissue and red represents relatively soft tissue. Real-time display of applied compression measured by a force sensor is depicted in the lower part of the image

Full size image

pSWE

In pSWE, malignancy shows higher SWS values than BPH and normal prostate tissue(2.37, 1.98, and 1.34 m/s, respectively) [68]. pSWE can be used to differentiate BPH and malignancy with an AUROC value of 0.86. SWS differences between the transition and peripheral zones of the prostate are possible in both BPH and cancer. When compared to DRE, pSWE shows higher diagnostic accuracy to detect malignancy, with AUROC value of 0.86 (vs. 0.67 for DRE) [68].

2D-SWE

2D-SWE has been shown to be useful for differentiating benign and malignant lesions in the peripheral zone [69]. Using a cut-off stiffness value of 35 kPa to differentiate benign and malignant lesions yields sensitivity and specificity values of 96% and 85%, respectively [70]. 2D-SWE is a reproducible technique with ICC value of 0.876 [71]. In a recent meta-analysis with 7 studies, Sang et al. reported pooled sensitivity and specificity values of 0.844 (95% CI 0.69–0.92) and 0.86 (95% CI 0.79–0.908) (AUROC value of 0.91) to differentiate malignant prostate lesions [72]. A different research group, Woo et al., reported similar results in their recent meta-analysis with 8 studies, pooled sensitivity value of 0.83 (95% CI 0.66–0.92), and specificity value of 0.85 (95% CI 0.78–0.9) [73]. 2D-SWE can also be used to assess BPH. Unlike most prostate malignancies, BPH develops from the transition zone of the prostate. Stiffness values of the transition zone can be measured via transrectal elastography. 2D-SWE can diagnose BPH with an AUROC value of 0.826 (95% CI 0.717–0.934) [74].

Thyroid

Strain elastography

Strain elastography requires external manual compression or physiological motion such as carotid pulsation [75]. The reported sensitivity of different strain imaging features for diagnosis of thyroid carcinoma ranges from 82% to 100%, with specificity ranging from 81.1% to 100% [76]. Although most studies indicate higher accuracy for thyroid cancer detection with strain elastography than conventional gray scale US, there is presently insufficient agreement among research groups regarding diagnostic criteria, and elastography is thought to be insensitive to some malignant tumor types [61, 76,77,78].

pSWE

Malignant thyroid nodules show higher SWS values when compared to benign nodules, either with pSWE or 2D-SWE [61]. pSWE can differentiate benign and malignant thyroid nodules. In a meta-analysis with 16 studies, pSWE has been reported to have an overall AUROC value of 0.91 [79]. In the assessment of diffuse chronic thyroid disease, pSWE is also useful to differentiate subjects with Graves’ disease and autoimmune thyroiditis from healthy subjects [80]. However, this is based on preliminary results and more studies are needed.

2D-SWE

2D-SWE is an effective technique to diagnose thyroid malignancies with AUROC values of 0.73 in nodules < 10 mm, 0.88 in nodules 11–30 mm, and 0.82 in nodules > 30 mm [81]. 2D-SWE has also been shown to potentially be able to differentiate benign and malignant follicular thyroid nodules, a clinically relevant finding that cannot be accomplished with FNA [15]. Figure 8 depicts 2D-SWE of normal thyroid tissue. Use of elastographic methods in combination with B-mode ultrasound is recommended [82]. Addition of CEUS can also increase diagnostic performance [83].

Fig. 8
figure 8

2D-SWE image of thyroid. Quantitative  SWE value is indicated on right side of the image

Full size image

Pancreas

Either strain or shear wave elastography can be used in the evaluation of the pancreas. Strain elastography is performed by using an endoscopic ultrasound system, in which aortic pulsation is used as the excitation method. Strain elastography with endoscopic ultrasound is limited by invasiveness, inadequate quality of images from the head and tail of the pancreas, and atherosclerotic changes that can affect aortic pulse excitation. Despite these limitations strain elastography has been used in a number of research studies to assess malignancy and pancreatic parenchymal disease [84].

Strain elastography

In a recent study of 149 patients, Rustemovic et al. proposed a strain ratio cut-off value of 7.59 to distinguish malignancies[100% sensitivity, 95% specificity] [85]. However, previous studies reported different results, which may be caused by the operator dependence of endoscopic ultrasound [84]. In a large mixed study population with 555 subjects that included healthy subjects, patients with chronic pancreatitis and patients with pancreatic cancer, Kim et al. reported mean strain ratio values of 3.78, 8.21, and 21.8, respectively. Sensitivity and specificity to distinguish malignancy were similar as Rustemovic et al.’s results [95.6% and 96.3%, respectively, cut-off 8.86] [86]. In their study with 191 patients with chronic pancreatitis, Iglesias-Garcia et al. reported sensitivity and specificity values of 91.2% and 91%, to detect chronic pancreatitis [AUROC value of 0.949], which are higher than Kim et al.’s values [87].

pSWE

pSWE can be used with cutaneous approach and it can detect stiffness differences between a lesion and background pancreatic parenchyma [88]. D’Onofrio et al. reported significant differences between SWS values obtained in adenocarcinoma and normal pancreatic parenchyma [89]. Patients with chronic pancreatitis may show higher stiffness values with pSWE when compared to healthy controls [4.3 vs. 2.8 kPa] [90]. Current knowledge of pancreatic elastography is limited to strain elastography and pSWE. Studies comparing elastography methods in the diagnosis of pancreatic masses and parenchymal diseases are also limited [91].

Spleen

Portal hypertension and increased hepatic venous pressure gradient (HVPG) are indicators of end-stage chronic liver disease, with increased risk of variceal bleeding, ascites variceal bleeding, ascites, and hepatic encephalopathy which increase mortality rates [2]. Accurate and fast detection of portal hypertension and esophageal varices, is critical to prevent complications, in which elastography techniques may fulfill this need. As mentioned above, strain elastography needs an excitation/pressure method. The spleen is in a relatively deep subcostal location, limiting utility of strain elastography for splenic assessment [61].

Transient elastography

Transient elastography probe can be performed in the spleen in a similar manner to the liver. It is known that TE can successfully detect stiffness of spleen in patients with cirrhosis, which correlates with HVPG which can be predictive for esophageal varices. Spleen stiffness value 3.3 m/s has been proposed as a cut-off value to rule out esophageal varices; however, more studies are needed [92]. In studies comparing TE application in liver and spleen to diagnose portal hypertension, diagnostic performance of liver TE in liver was reported to be higher than splenic TE (AUROC’s 0.95 vs. 0.85, respectively) [2]. Use of transient elastography in combination with conventional ultrasound may help operators to locate the most reliable location [61].

pSWE

pSWE has also been studied in cirrhotic patients. To diagnose clinically significant portal hypertension,  pSWE technique was reported to have an AUROC of 0.943 and to detect presence of esophageal varices, pSWE technique was reported to have AUROC of 0.933 [93, 94]. Recently, research groups reported studies with 2D-SWE to predict esophageal varices and portal hypertension [95,96,97]. Elkrief et al. evaluated the performance of TE and 2D-SWE to detect portal hypertension by measuring liver stiffness and spleen stiffness, and reported higher AUROC value for liver stiffness measurements when compared to spleen stiffness(0.87 vs. 0.64). They also reported superior technical success rate of 2D-SWE when compared to TE, in assessment of liver and spleen stiffness [97]. Although proposed cut-off values of elastography methods to detect portal hypertension or esophageal varices through evaluation of spleen stiffness are similar, more studies are needed to verify these cut-off values.

In this brief review, we have discussed general principles of different elastography technologies and clinical applications of these methods. Summary and classification of all elastographic techniques are indicated in Fig. 9 [10, 11, 15, 16, 27, 31,32,33, 37,38,39, 49, 51, 57, 62, 64, 66, 68, 69, 76, 79, 89, 98]. Strong features and limitations of these techniques are summarized in Figs. 10 and 11 [2, 36, 61, 92, 99,94,95,96,97,98,100,101,102,103,104,105,106,107].

Fig. 9
figure 9

Summary and classification of elastography techniques

Full size image
Fig. 10
figure 10

Strong features of elastography techniques

Full size image
Fig. 11
figure 11

Limitations of elastography techniques

Full size image

Conclusion

Ultrasound elastography comprises a set of techniques that non-invasively measure tissue stiffness. Use of these techniques has blossomed with recognition that many disease processes affect tissue stiffness, providing a new imaging target for assessment of disease biology. In this review, we have provided a brief introduction to the physical concepts that underpin ultrasound elastography, and have discussed several different commercially available ultrasound elastography systems with evidence of their efficacy in different biologic settings. With the help of guidelines, meta-analysis reports, and studies with large study populations, various cut-off values are determined. However, there may be differences between measurements with different ultrasound systems. Researchers and clinicians should liaise with manufacturers regarding their proposed cut-off values for specific elastography applications.